© 2015. Published by The Company of Biologists Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License

(http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution and reproduction in any medium provided that the original work is properly attributed.

The Meckel-Gruber syndrome protein TMEM67 controls basal body

positioning and epithelial branching morphogenesis via the non-canonical

Wnt pathway

Zakia A. Abdelhamed1,2, Subaashini Natarajan1, Gabrielle Wheway1, Christopher F.

Inglehearn1, Carmel Toomes1, Colin A. Johnson1*, Daniel J. Jagger3*

1 Ciliopathy Research Group, Section of Ophthalmology and Neurosciences, Leeds Institute

of Molecular Medicine, University of Leeds, Leeds, LS9 7TF, UK.

2 Department of Anatomy & Embryology, Faculty of Medicine, Al-Azhar University, Cairo,

Egypt.

3 UCL Ear Institute, University College London, 332 Gray’s Inn Road, London WC1X 8EE,

UK.

* Corresponding authors:

Dr. Daniel J. Jagger, UCL Ear Institute, University College London, 332 Gray’s Inn Road,

London WC1X 8EE, UK

e-mail d.jagger@ucl.ac.uk tel: +44 (0)207 679 8930 fax: +44 (0)207 679 8990

Prof. Colin A. Johnson, Department of Ophthalmology and Neurosciences, Leeds Institute of

Molecular Medicine, Wellcome Trust Brenner Building, St. James’s University Hospital,

Leeds, LS9 7TF, UK

e-mail c.johnson@leeds.ac.uk tel: +44 (0)113 343 8443 fax: +44 (0)113 343 8603

Keywords: TMEM67, meckelin, MKS3, Wnt signalling, planar cell polarity, PCP,

stereocilia; kinocilia; primary cilia; hair bundle; ciliopathy

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

http://dmm.biologists.org/lookup/doi/10.1242/dmm.019083Access the most recent version at DMM Advance Online Articles. Posted 7 April 2015 as doi: 10.1242/dmm.019083http://dmm.biologists.org/lookup/doi/10.1242/dmm.019083Access the most recent version at

First posted online on 7 April 2015 as 10.1242/dmm.019083

Abstract

Ciliopathies are a group of developmental disorders that manifest with multi-organ

anomalies. Mutations in TMEM67 (MKS3) cause a range of human ciliopathies,

including Meckel-Gruber and Joubert syndromes. In this study we describe multi-

organ developmental abnormalities in the Tmem67tm1Dgen/H1 knockout mouse that closely

resemble those of Wnt5a and Ror2 knockout mice. These include pulmonary hypoplasia,

ventricular septal defects, shortening of the body longitudinal axis, limb abnormalities,

and cochlear hair cell stereociliary bundle orientation and basal body/kinocilium

positioning defects. The basal body/kinocilium complex was often uncoupled from the

hair bundle, suggesting aberrant basal body migration, although planar cell polarity

and apical planar asymmetry in the organ of Corti were normal. TMEM67 (meckelin) is

essential for phosphorylation of the non-canonical Wnt receptor ROR2 (receptor

tyrosine kinase-like orphan receptor 2) upon Wnt5a stimulation. ROR2 also co-localizes

and interacts with TMEM67 at the ciliary transition zone. Additionally, the

extracellular N-terminal domain of TMEM67 preferentially binds to Wnt5a in an in

vitro binding assay. Tmem67 mutant lungs in ex vivo culture failed to respond to Wnt5a

stimulation of epithelial branching morphogenesis. Wnt5a also inhibited both the Shh

and canonical Wnt/β-catenin signalling pathways in normal embryonic lung.

Pulmonary hypoplasia phenotypes, including loss of correct epithelial branching

morphogenesis and cell polarity, were rescued by stimulating the non-canonical Wnt

pathway downstream of the Wnt5a-TMEM67-ROR2 axis by activating RhoA. We

propose that TMEM67 is a novel receptor that has a major role in non-canonical Wnt

signalling, mediated by Wnt5a and ROR2, and normally represses Shh signalling.

Downstream therapeutic targeting of the Wnt5a-TMEM67-ROR2 axis could reduce or

prevent pulmonary hypoplasia in ciliopathies and other congenital conditions.

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

Introduction

Primary cilia are microtubule-based organelles that sense and transduce extracellular

signals on many mammalian cell types. The cilium is known to play essential roles

throughout development in mechanosensation (Praetorius and Spring, 2001; Nauli et al.,

2003), in signal transduction by the Hedgehog, Wnt and PDGFRα signalling pathways

(Huangfu et al., 2003; Simons et al., 2005; Schneider et al., 2005) and in the establishment of

left-right asymmetry (Nonaka et al., 1998). Primary cilia have a complex ultrastructure with

compartmentalization of molecular components that combine in functional modules.

Components that are required for both the formation and function of the cilium have to be

transported from the cytoplasm of the cell by the process of intraflagellar transport (IFT).

Mutations in proteins that are structural or functional components of the primary cilium cause

a group of human inherited conditions known as ciliopathies (Adams et al., 2008). The loss of

these components can disrupt ciliary functions such as the control of protein entry and exit

from the cilium, the possible trafficking of essential ciliary components, and the regulation of

signalling cascades and control of the cell cycle. Many proteins that are mutated in

ciliopathies are localized to the transition zone, a compartment of the proximal region of the

cilium (Szymanska and Johnson, 2012; Reiter et al., 2012). In particular, a protein complex at

the transition zone, known as the “MKS-JBTS module”, contains many of the proteins

mutated in Meckel-Gruber syndrome (MKS) and Joubert syndrome (JBTS) (Garcia-Gonzalo

et al., 2011; Sang et al., 2011).

MKS is the most severe ciliopathy, and is a lethal recessive neurodevelopmental

condition. The central nervous system (CNS) defects often comprise occipital encephalocele,

rhombic roof dysgenesis and prosencephalic dysgenesis. Cystic kidney dysplasia and hepatic

developmental defects are essential diagnostic features of MKS, and although the CNS

defects are considered to be obligatory features they have a more variable presentation. Other

occasional features include post-axial polydactyly, shortening and bowing of the long bones,

retinal colobomata and situs defects. To date, mutations in eleven genes have been described

as a cause of MKS. However, mutations in the TMEM67/MKS3 gene are the most common

cause of MKS, accounting for over 15% of all MKS cases in unselected cohorts (Khaddour et

al., 2007; Consungar et al., 2007; Szymanska et al., 2012), with mutations in TMEM67

associated frequently with a diagnosis of ductal plate malformation in the liver (Khaddour et

al., 2007; Consungar et al., 2007; Szymanska et al., 2012). TMEM67 encodes TMEM67

(transmembrane protein 67, also known as meckelin), a 995 amino-acid transmembrane

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

protein with structural similarity to Frizzled receptors (Smith et al., 2006).

TMEM67/meckelin (hereafter called TMEM67) contains an extracellular N-terminal domain

with a highly conserved cysteine-rich repeat domain (CRD), a predicted β-pleated sheet

region and seven predicted transmembrane regions (Abdelhamed et al., 2013). TMEM67 is a

component of the MKS-JBTS module at the transition zone. This functional module includes

other transmembrane proteins, namely the Tectonic proteins (TCTN1 to 3), TMEM17,

TMEM231 and TMEM237, as well as C2-domain proteins (jouberin/AHI1 and CC2D2A)

(Sang et al., 2011; Garcia-Gonzalo et al., 2011; Huang et al., 2011; Chih et al., 2011).

Transition zone proteins are thought to form a diffusion barrier at the base of the cilium that

restricts entrance and exit of both membrane and soluble proteins (Williams et al., 2011;

Garcia-Gonzalo et al., 2012).

Loss or dysfunction of cilia in MKS causes complex de-regulation of key normal

pathways of embryonic development such as Wnt and Shh signalling (Abdelhamed et al.,

2013). In particular, primary cilia have been proposed to mediate a negative modulatory

effect on the canonical Wnt/β-catenin pathway (Simons et al., 2005; Gerdes et al., 2007;

Corbit et al., 2008; Lancaster et al., 2011). In contrast, less is known about the possible

regulatory roles of cilia and ciliary compartments on the non-canonical pathways of Wnt

signalling. Downstream effects of non-canonical Wnt signalling (also referred to as planar

cell polarity or PCP) result in cytoskeletal-actin rearrangements that cause changes in cell

morphology and their directed orientation relative to a planar axis within an epithelium. Actin

cytoskeleton remodelling is mediated by Rho proteins, a family of small GTPases that

regulate many aspects of intracellular actin dynamics. In vertebrates, PCP signalling is

required for correct convergent extension (Jessen et al., 2002; Ybot-Gonzalez et al., 2007)

which, when disrupted, can cause neural tube defects, misorientation of hair cells and

disruption of stereociliary bundles in the mammalian cochlea (Montcouquiol et al., 2003),

and misorientation of hair follicles in the epidermis (Devenport and Fuchs, 2008). The

importance of cilia for PCP signalling has been shown for ciliary proteins (namely, certain

Bbs proteins and Ift88) that are required for the correct regulation of basal body polarization

in the cochlea (Ross et al., 2005; Jones et al., 2008). Furthermore, the core PCP protein

Dishevelled (Dvl) and other core PCP proteins (such as Dubroya, Frizzled and Celsr2/3) are

involved in the assembly and remodelling of the actin cytoskeleton in apical cellular regions

(Oishi et al., 2006; Valente et al., 2010; Tissir et al., 2010), allowing subsequent ciliogenesis

by the docking basal bodies to the apical cellular membrane (Park et al., 2008). Consistent

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

with a role in non-canonical Wnt signalling, TMEM67 is required for centriolar migration to

the apical membrane (Dawe et al., 2007), as well as regulation of actin cytoskeleton

remodelling and RhoA activity (Dawe et al., 2009). Furthermore, Wnt5a (an activator of non-

canonical but an inhibitor of the canonical Wnt pathway) stimulated the aberrant formation of

extensive actin stress fibres in the absence of TMEM67 (Abdelhamed et al., 2013). However,

the role of TMEM67 in non-canonical Wnt signalling or the PCP signalling system is

unknown, and it remains undetermined if TMEM67 binds to the Wnt5a ligand or is essential

for co-receptor function.

To begin to answer these questions, the present study focuses on PCP and non-

canonical Wnt signalling defects in the recently characterized Tmem67tm1(Dgen/H) knockout

mouse (Abdelhamed et al., 2013; Garcia-Gonzalo et al., 2011), hereafter referred to as the

Tmem67-/- knockout mutant. We now show that the pulmonary and cardiological phenotypes

of Tmem67-/- mutant embryos closely recapitulate those of Wnt5a and Ror2 mutant mice

(Oishi et al., 2003). To substantiate a possible role of TMEM67 in the non-canonical Wnt

signalling pathway, we examined the morphogenesis of the cochlea in neonatal Tmem67-/-

mice, a well-characterized model system to determine PCP defects in a developing embryo

(Jones and Chen, 2007). Analysis of the orientation of stereociliary hair bundles, and the

positioning of primary cilia and basal bodies, demonstrated a consistent TMEM67-dependent

effect on cochlear PCP. We then used biochemical methods to show the domains of

interaction between TMEM67 and either Wnt5a or ROR2 (receptor tyrosine kinase-like

orphan receptor 2), a non-canonical Wnt receptor. We also functionally characterized the

response of lung tissue explanted ex vivo for external Wnt5a stimulation, showing that

normal epithelial branching morphogenesis and cell polarity was lost in the absence of

TMEM67 but could be rescued by activation of RhoA. Our results suggest that TMEM67 is a

novel receptor/co-receptor of non-canonical Wnt signalling that preferentially binds Wnt5a

with the extracellular cysteine-rich domain (CRD) and mediates downstream signalling

through ROR2 as a co-receptor. TMEM67 could therefore be essential for ROR2 function

and the correct activation of downstream non-canonical Wnt signalling cascades.

Results

Tmem67-/- embryos recapitulate the phenotypes of Wnt5a and Ror2 knockout animals

The majority of mutant Tmem67-/- pups died at birth, and none lived beyond the

second postnatal day (P1), most likely because of pulmonary hypoplasia and complex cardiac

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

malformations that include ventricular septal defect (VSD). Both phenotypes were consistent

anomalies detected in Wnt5a and Ror2 mutant animals. Morphological and histological

examination of Tmem67 mutants showed that the lungs were hypoplastic (Figure 1A) with

failure of the pulmonary alveoli to develop (Figure 1B-C). Interstitial cells also had

increased cell proliferation as determined by staining for the proliferation marker Ki-67

(Figure 1B). Primary cilia were significantly reduced in both length and number on cells

forming the pulmonary alveoli and distal air sacs in Tmem67-/- embryonic lungs (Figure 1C).

Limb dysplasia, omphalocele and intrauterine growth retardation were detected in

20% (n=4/20) of Tmem67-/- embryos (Figure 1D). Caudal truncation with a shortened

anterior-posterior axis was detected in 60% of mutant pups (n=12/20) (Figure 1D). A small

proportion of E11.5 Tmem67-/- embryos (n=1/12) developed an inverted tail turning (Figure

1A), the earliest sign of laterality defects. Later in development at the perinatal (E15.5) and

early postnatal stages (P0), 100% (n=7/7) of investigated mutant animals had left pulmonary

isomerism (Figure 1A). Both the right and left lungs appeared indistinguishable from each

other and were formed of two identical symmetrical lung lobes. In the Tmem67+/+ wild-type

embryos, the right and left lungs were easily differentiated by the identification of four and

one lobes in the right and left lungs, respectively (Figure 1A).

Cardiac oedema consistently developed in most of the animals analysed. Complex

cardiac developmental defects, including ventricular septal defect, atrial septal defect and

dextrocardia, were common malformations detected in Tmem67-/- embryos (n=6/8) (Figure

1E-F). All mutant Tmem67-/- embryos had evidence of a ductal plate malformation and the

retention of multiple primitive bile duct structures (Figure 1G), consistent with the hepatic

developmental anomalies observed in human patients with TMEM67 mutations (Adams et al.,

2008; Khaddour et al., 2007; Consungar et al., 2007; Szymanska et al., 2012) and in the

Wnt5a and Ror2 mutant mice (Kiyohashi et al., 2013). The pulmonary, cardiological and

hepatic phenotypes of Tmem67-/- mutant embryos therefore closely recapitulate those of

Wnt5a and Ror2 mutant mice (Oishi et al., 2003). In addition, the caudal truncation and

shortened anterior-posterior axis in P0 Tmem67-/- mutant pups is similar to that of Wnt5a

knock-out mice.

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

Cochleae of neonatal Tmem67-/- mutants display abnormalities of hair bundle

orientation with uncoupling of primary cilia and basal bodies, but have normal planar

cell polarity and apical planar asymmetry

To further investigate the possible role of TMEM67 in the non-canonical Wnt

signalling pathway, we examined the morphogenesis of the cochlea in neonatal Tmem67-/-

mice. We mapped the distribution of TMEM67 in the neonatal organ of Corti, and analysed

the orientation of stereociliary hair bundles and the position of primary cilia to determine

TMEM67-dependent effects on cochlear planar cell polarity (PCP). Cochleae from P0

Tmem67-/- mice were normal in appearance and comparable in size to those of littermate

controls (Figure 2A). Phalloidin staining of whole-mount preparations of the organ of Corti

(the sensory neuroepithelium) revealed that the total epithelial length was not different

between the genotypes (Figure 2B), suggesting that TMEM67 does not play a direct role in

the PCP-associated convergent extension mechanisms that underlie growth of the organ of

Corti along the baso-apical axis (Dabdoub and Kelley, 2005). The organ of Corti, which is

shown in schematic form in Figure 2C, is an epithelial mosaic housing a single row of inner

hair cells (ihc) and generally three rows of outer hair cells (ohc), which are interspersed with

non-sensory supporting cells. During normal development, all cells in the epithelium possess

a single cilium that projects from their apical (luminal) surface, whilst hair cells can be

identified by their actin-containing stereociliary bundles. TMEM67 was localised to the

proximal regions of acetylated α-tubulin-stained cilia of hair cells and the supporting cells of

P0 wild-type mice (Figure 2D), consistent with its previously described localization to the

ciliary transition zone (Simons et al., 2005; Garcia-Gonzalo et al., 2011).

Along the whole baso-apical axis of both Tmem67+/+ and Tmem67-/- cochleae there

was a single continuous row of ihc located along the neural (medial) edge of the sensory

epithelium (Figure 2E). Similarly, there were three continuous rows of ohc running parallel

to the abneural (lateral) edge in all animals. The normal cochlear morphogenesis further

suggests that TMEM67 does not contribute to cochlear convergent extension. The phalloidin-

stained hair bundles of Tmem67+/+ ihc and ohc were all regularly oriented (Figure 2E), with

the vertex of the “V-shaped” bundle generally directed towards 0° (the abneural pole; Figure

2C). Similarly, the stereociliary hair bundles of ihc in neonatal Tmem67-/- mice had a regular

orientation. However, there were marked abnormalities in the alignment of ohc stereociliary

hair bundles in neonatal Tmem67-/- mice, a phenotype that was most noticeable in the basal

cochlear turn, where ca. 30% of place-matched ohc had misoriented bundles relative to the

abneural pole. Misoriented ohc often retained a roughly V-shaped hair bundle (Figure 2E;

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

Tmem67-/- basal turn, inset). In the apical (least mature) regions, the ohc bundle abnormalities

were still apparent but had a lower occurrence.

Primary cilia were detected on the surface of hair cells and non-sensory supporting

cells in the basal cochlear region of Tmem67+/+ mice (Figure 2E). The primary cilia of hair

cells (known as “kinocilia”) were all located close to the vertex of the regularly aligned hair

bundles. Kinocilia were also detected on the surface of all Tmem67-/- hair cells, and these

were located in approximately normal positions on ohc with hair bundles oriented towards 0°,

and on some ihc. On ohc with noticeably misorientated hair bundles, the kinocilium was

eccentrically localized, and consequently found mis-positioned relative to the bundle vertex.

In such instances, the kinocilium rarely appeared to contact the tallest row of stereocilia at the

rear of the bundle. In most ihc, although the hair bundle was oriented normally, kinocilia

were positioned eccentrically and were not attached to the hair bundle. There was an absence

of cilia on supporting cells in the lateral portion of the organ of Corti of Tmem67-/- mutants,

namely the Deiters’ cells and outer pillar cells.

The uncoupling of cochlear cilia from hair bundles in neonatal Tmem67-/- mutants was

further investigated by a quantitative analysis of the basal body position in ohc and ihc along

the baso-apical axis of the organ of Corti (Figure 2F-G), since the localization of the basal

body has been used as a measure of the PCP axis in hair cells (Jones and Chen, 2007). The

basal body in hair cells could be delineated by the anti-ALMS1 antibody (Figure 2F),

allowing the precise measurement of position relative to 0°. Scatter plots of hair bundle

orientation versus basal body position for individual basal turn ohc demonstrated the

variation of the uncoupling defect in Tmem67 mutants (Figure 2G). In Tmem67+/+ hair cells

there was close correlation between hair bundle orientation and basal body position

(Pearson’s coefficient of correlation, r = 0.86). For Tmem67-/- mutant ohc, although some

cells had close coupling of the basal body and hair bundle, there was an overall broader

overall distribution (r = 0.71). An analysis of the average deviation of the basal body position

from 0° (Figure 2H) revealed that there was significant mis-localization in each row of

Tmem67-/- hair cells along the mutant cochleae, and that there was a place-dependent

variability within the medio-lateral axis. In contrast, the positional deviation of basal bodies

in Tmem67+/+ hair cells was identical to previous measurements of hair bundle orientation at

this gestational age (Jones and Chen, 2007). Distribution histograms for basal body position

in hair cells (Suppl. Figure 1) further demonstrated the variability of the mis-localisation

along the baso-apical and medio-lateral axes of Tmem67-/- mutant cochleae. In contrast, both

planar cell polarity and apical planar asymmetry were undisturbed in the organ of Corti of

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

neonatal Tmem67-/- mice, by IF staining for the core PCP protein Vangl2 (Montcouquiol et

al., 2003), and the asymmetrically localized GTP-binding protein alpha-i subunit 3 (G i3)

and atypical Protein Kinase C (aPKC; Ezan et al., 2013) (Figure 3).

Basal body mis-localization defects during hair cell differentiation in embryonic

Tmem67-/- mutants

To further investigate the ontogeny of the basal body mis-positioning in Tmem67-/-

mutant hair cells during late gestation, we examined the sensory epithelium during a prenatal

period when hair cells and supporting cells begin to differentiate within the pro-sensory

domain. The cell types can be distinguished first in the basal region between E14 and E15 in

the mouse cochlea, and then along the whole baso-apical axis by E17 (Dabdoub and Kelley,

2005). At E15.5, hair cells could be clearly defined by phalloidin staining in the basal

cochlear region (Suppl. Figure 2). In the basal region, primary cilia were detected on

Tmem67+/+ hair cells and supporting cells (Suppl. Figure 2A) but, as observed in P0

animals, Tmem67-/- supporting cells in the lateral region lacked primary cilia (Suppl. Figure

2A). The kinocilium appeared centrally on the apical surface of a hair cell, and subsequently

migrated to the abneural pole (Jones et al., 2008). In the basal turn of E15.5 Tmem67+/+ mice,

ALMS1-labelled basal bodies had already migrated to the abneural pole in ihc and rows 1-2

of ohc (Suppl. Figure 2B). In Tmem67-/- mutant littermates, ihc basal bodies appeared to

have a largely normal localisation, but basal bodies of ohc in all rows were often found

centrally or had apparently migrated eccentrically towards the cell periphery (Suppl. Figure

2B-C). This suggests that TMEM67 regulates the migration of ohc basal bodies towards the

cell periphery but not those of ihc, and may specify the final position of basal bodies in all

hair cells relative to 0°. In the mid-turn region of both genotypes, ihc had polarised basal

bodies but basal bodies in all ohc rows were centrally located (Suppl. Figure 2B-C),

suggesting migration had yet to commence at this less developed region of the baso-apical

axis.

TMEM67 is required for negative regulation of the canonical Wnt/ -catenin signalling

pathway by Wnt5a and interacts with ROR2

We next used biochemical methods to substantiate that Tmem67-/- cells have a defect

in the regulation of non-canonical Wnt signalling that is concomitant with loss of negative

modulation of the canonical Wnt/β-catenin pathway. TMEM67 is a putative orphan receptor

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

with similarities to the Frizzled proteins (Figure 4A) (Smith et al., 2006; Abdelhmaed et al.,

2013), and we therefore next used the TOPFlash assay to quantify the ability of Tmem67+/+

and Tmem67-/- mouse embryonic fibroblasts (MEFs) to respond to Wnt ligands. After co-

transfection of the TOPFlash reporter constructs, treatment with Wnt3a stimulated basal

levels of Wnt/ -catenin signalling by about 5-fold in Tmem67+/+ MEFs, but by 13.8-fold in

mutant cells (Figure 4B). Co-transfection with a wild-type TMEM67 construct completely

rescued a normal response in Tmem67-/- MEFs by suppressing the deregulated canonical

Wnt/β-catenin signalling responses to Wnt3a (Figure 4B). However, TMEM67 constructs

with the pathogenic missense mutations M252T, L349S, Q376P and R440Q in the

extracellular N-terminal (Nt) domain of TMEM67 (Figure 4A) were unable to restore normal

basal levels of canonical Wnt/ -catenin signalling (Figure 4B). Two other pathogenic

missense mutations, R549C and C615R, located close to transmembrane helices (Figure 4A),

also did not rescue basal responses to Wnt3a (Figure 4B). Although Wnt5a on its own had no

effect on the canonical pathway (Abdelhamed et al., 2013), treating cells with a mixture of

Wnt3a and Wnt5a showed that the latter ligand was able to competitively inhibit the Wnt3a

response in wild-type cells, but only partially inhibited the Wnt3a response in mutant cells. In

Tmem67-/- cells, the missense mutations in the extracellular Nt domain of TMEM67 did not

rescue the competitive inhibition of Wnt3a canonical responses by Wnt5a (Figure 4C). Wild-

type TMEM67 partially rescued the correct response as expected (Figure 4C), implying that

Wnt5a modulates a non-canonical Wnt signalling response through TMEM67.

Since the cardiological and pulmonary phenotypes of Tmem67-/- mutant embryos

(Figure 1A-C, E-F) closely recapitulate those of Wnt5a and Ror2 mutant mice and P0 pups

exhibit a shortened anterior-posterior axis (Figure 1D) similar to Wnt5a knock-out mice, we

hypothesized that TMEM67 could be a potential receptor that directly binds Wnt ligands. To

test this, we performed an in vitro binding assay using purified, fluorescein-labelled Wnt3a or

Wnt5a proteins (Figure 4D). Titration with increasing amounts of wild-type TMEM67-Nt

protein (Figure 4D), demonstrated a preferential binding to Wnt5a compared to Wnt3a

(Figure 4E). Missense mutations (M252T, L349S, Q376P and R440Q) in the extracellular N-

terminal region of TMEM67 (Figure 4A) completely abolished binding to Wnt5a (Figure

4F). We were, however, unable to test the TMEM67-Nt R549C and C615R proteins because

the proximity of hydrophobic residues in the transmembrane helices prevented efficient

protein expression (data not shown).

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

ROR2 (receptor tyrosine kinase-like orphan receptor 2) is known to mediate non-

canonical Wnt5a signalling (Mikels et al., 2009). Next, we therefore investigated the possible

functional interactions between ROR2 and TMEM67. Endogenous ROR2 co-localized with

both TMEM67 and RPGRIP1L, a marker of the transition zone (Arts et al. 2007), in ciliated

mIMCD3 cells (Figure 5A). Consistent with this observation, exogenously expressed FLAG-

tagged ROR2 also partially co-localized with endogenous ROR2 and TMEM67 and in

ciliated mIMCD3 cells (Suppl. Figure 3A), and with -tubulin at the base of primary cilia in

Tmem67+/+ wild-type and Tmem67-/- mutant MEFs (Suppl. Figure 3B). Co-

immunoprecipitation experiments demonstrated that exogenous full-length and endogenous

TMEM67 interacted with FLAG-tagged ROR2 (Figure 5C & 5D) but not a tagged irrelevant

protein (MCPH1). We then confirmed non-canonical Wnt pathway dysregulation in the

absence of TMEM67 by transfecting MEFs with FLAG-ROR2. As expected, levels of the

activated phosphorylated ROR2 isoform were significantly increased following treatment of

wild-type Tmem67+/+ MEFs with Wnt5a, but active ROR2 was completely abolished in the

mutant Tmem67-/- cells (Figure 5E).

Defective branching morphogenesis in response to Wnt5a stimulation in the Tmem67-/-

embryonic lung is rescued by the RhoA activator calpeptin

We reasoned that if TMEM67 is a potential receptor that directly binds to Wnt5a, the

absence of the receptor in the mutant would abolish or reduce responses to this ligand. We

therefore next used an ex vivo organogenesis assay to follow epithelial branching

morphogenesis in embryonic (E12.5) lung in response to Wnt5a. As expected, wild-type

Tmem67+/+ lung had a strong response to this Wnt ligand, in comparison to control

treatments, with prolific elaboration of distal branching in the developing alveoli (Figure 6A

& B, Suppl. Figure 4). Consistent with the pulmonary phenotypes of Tmem67-/- mutant

embryo (Figure 1A-B), Tmem67-/- mutant lungs grown in ex vivo culture were hypoplastic

with significantly reduced levels of branching (Figure 6A & B). Mutant lungs did not

respond to treatment with Wnt5a, consistent with a role for TMEM67 in binding Wnt5a

during embryonic processes such as hair cell differentiation and lung morphogenesis.

Consistent with a loss of responsiveness to non-canonical Wnt signalling, we observed

reduced levels of active RhoA in embryonic (E14.5) Tmem67-/- mutant lung (Figure 6C). In

contrast, expression of Shh and downstream effectors of the Shh pathway (Gli1 and Ptch1)

were significantly increased in embryonic Tmem67-/- mutant lung (Figure 6D). Consistent

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

with previous studies (Abdelhamed et al., 2013; Garcia-Gonzalo et al., 2011), canonical Wnt

signalling, as measured by Axin2 expression, was also increased in mutant lung (Figure 6D).

In the absence of TMEM67, ROR2 phosphorylation is therefore lost and the normal

regulation of non-canonical Wnt signalling is disrupted. We reasoned that activation of a

more downstream target of this pathway could potentially enhance lung maturation and

rescue the abnormal branching, mimicking the correct responses to Wnt5a. To test this

hypothesis, we used the ex vivo organogenesis assay to treat embryonic (E15.5) wild-type

Tmem67+/+ and mutant Tmem67-/- lungs with calpeptin. Calpeptin is a dipeptide aldehyde that

inhibits myosin light chain phosphorylation connected to stress fibre formation, specifically

targeting regulators of the Rho sub-family of GTPases and selectively activating RhoA

(Schoenwaelder and Burridge, 1999; Schoenwaelder et al., 2000). Mutant lungs at embryonic

ages E11.5 and E13.5 showed areas of delayed and abnormally dilated branches surrounded

by areas of condensed mesenchyme (Figure 7A, Suppl. Figure 5A). Treatment with

calpeptin resulted in the appearance of more developed branches and less condensed

mesenchyme, closely resembling the morphology of wild-type lung, at both E11.5 and E13.5

(Figure 7A-B, Suppl. Figure 5A). Histological assessment of these developmental changes

after calpeptin treatment showed that Tmem67-/- lungs at E13.5 had more developing alveoli

and greatly reduced the mesenchymal cell condensations, with maturation comparable to

wild-type lungs (Suppl. Figure 5B). In wild-type Tmem67+/+ embryonic lungs, the

orientation of mitotic division in alveolar epithelial cells was predominately perpendicular to

the apical cell surface and basement membrane (Figure 7C). In mutant Tmem67-/- alveoli,

mitotic divisions were predominantly parallel, but treatment with calpeptin rescued normal

polarity (Figure 6C).

Discussion

We have previously described the severe multi-organ developmental defects in the

B6;129P2-Tmem67tm1Dgen/H knock-out mouse, which reiterate the clinical features of MKS

and Joubert syndrome (JBTS) (Abdelhamed et al., 2013). All Tmem67-/- mutants that were

examined, developed incomplete laterality defects that manifested in later gestation as left

lung isomerism (Figure 1A) and were occasionally associated with dextrocardia (Figure 1E-

F). Pulmonary hypoplasia was a consistent finding in the Tmem67-/- embryos and pups

(Figure 1A-B), although this is infrequently under-reported in human ciliopathies and is not

considered as an essential diagnostic clinical feature of MKS in humans (Salonen, 1984).

However, it is been reported recently that most MKS patients are either stillborn or die within

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

hours after birth because of the pulmonary hypoplasia, and this can be considered as the

leading cause of death in human MKS patients (Roy and Pal, 2013).

Previously, we have shown that TMEM67 is required for epithelial branching

morphogenesis in three-dimensional in vitro tissue culture (Dawe et al., 2007). The present

study now provides the first evidence that TMEM67 is essential for correct in vivo branching

morphogenesis in lung alveolar system development (Figure 6A & B). The similarity in the

overall cardiopulmonary phenotypes (Oishi et al., 2003) and the biliary developmental

malformations (Kiyohashi et al., 2013) for Wnt5a, Ror2 and Tmem67 knock-out mice

(Figure 1) strongly suggests that TMEM67 mediates signalling by either the Wnt5a ligand or

the ROR2 co-receptor. A marked phenotype of Wnt5a-/- mice is convergent-extension defects

with mis-orientation of ohc and ihc stereociliary bundles (Qian et al., 2007). To further test if

Wnt5a signals through TMEM67, we therefore investigated the morphogenesis of the cochlea

in neonatal Tmem67-/- mice.

In the present study, we now show that TMEM67 is a key regulator of cilium-

dependent stereociliary hair bundle orientation. In Tmem67 mutant mice, ohc had mis-

oriented hair bundles (Figure 2E) with an apparent physical dissociation of the basal

body/kinocilium complex from the hair bundle (Figure 2F-G). This uncoupling may arise

from aberrant migration of the basal body, during a period of embryonic development

immediately prior to the initial growth of the stereocilia (Suppl. Figure 2). In mutant ihc, the

basal body migrated towards the abneural pole of the cell, but the fine control of its final

positioning appeared to be variable. These results are consistent with our previous work that

has implicated TMEM67 in mediating centriole migration to the apical membrane of

polarized cells with the consequent formation of a primary cilium (Dawe et al., 2007).

TMEM67 also contributed to ciliogenesis in the organ of Corti, although this appeared to be

specific to the non-sensory supporting cells since all sensory hair cells were ciliated. This

observation is consistent with previous results in ciliated cell-lines (Dawe et al., 2007), in

other tissues of Tmem67-/- mutants (Adams et al. 2012; Abdelhamed et al., 2013), and in the

organ of Corti of the bpck (Tmem67 null) mouse (Leightner et al., 2013). The latter study also

reported stereociliary alignment and ciliogenesis defects in bpck mutant neonates, but did not

investigate basal body migration or positioning defects in embryos (Leightner et al., 2013).

The hair bundle orientation defects in both bpck and Tmem67-/- lines are similar to

those observed in mouse models of the human ciliopathies such as Alström syndrome (Jagger

et al., 2011), BBS (May-Simera et al., 2009; Ross et al., 2005), and the Kif3a ciliary mutant

(Sipe and Lu, 2011). Unlike Kif3a-/- mice, however, Tmem67 mutants had the expected

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

number of hair cell rows and the length of the sensory epithelium was comparable to that in

controls (Figure 2B & E), and both planar cell polarity and apical planar asymmetry were

normal (Figure 3) indicating that cochlear convergent-extension mechanisms were

unaffected by loss of TMEM67. In Kif3a-/- hair cells, basal body position shows little

correlation with the hair bundle orientation (Sipe and Lu, 2011), comparable to the

orientation defects observed in Tmem67 mutants (Figure 2E & G), suggesting that hair

bundle orientation does not necessarily predict the position of the basal body (Suppl. Figure

1, Suppl. Figure 2B). The basal body therefore appears to be a better assay of the PCP axis

(Sipe and Lu, 2011). Importantly, the Tmem67 model system also provides in vivo

confirmation of previous in vitro studies that suggested that TMEM67 has an essential role in

mediating centriolar migration to the apical membrane during cell polarization (Dawe et al.,

2007).

Our biochemical data also suggests that non-canonical Wnt signalling by Wnt5a is

mediated or regulated, at least in part, by TMEM67 through a ciliary-dependent mechanism.

In ex vivo cultured Tmem67-/- lungs, a reduction in the number of epithelial branches was

detected from E12.5 (Figure 6A). Wnt5a treatment failed to induce an increase in epithelial

branching in Tmem67-/- lungs whereas wild-type lungs responded to this treatment with

prolific branching morphogenesis (Figure 6A, Suppl. Figure 4), suggesting that Tmem67-/-

lungs are unresponsive to non-canonical Wnt5a stimulation. A proposed functional

interaction between Wnt5a, ROR2 and TMEM67 is supported by several lines of

experimental evidence: preferential in vitro binding of the TMEM67 CRD domain to Wnt5a

(Figure 4E); the co-localization and interaction of ROR2 with TMEM67 at the ciliary

transition zone (Figure 5A-C); and the failure of Tmem67-/- cells to phosphorylate ROR2

upon Wnt5a stimulation (Figure 5D).

ROR2 is a member of the receptor tyrosine kinase (RTKs) superfamily and the

cytoplasmic regions of the RTKs family contain conserved tyrosine kinase domains

(Robinson et al., 2000; Sossin, 2006; Green et al., 2008). Similar to other RTKs, ROR2 forms

homodimers at the cell membrane, an event essential for receptor trans-autophosphorylation

and subsequent pathway activation (Green et al., 2008; Kani et al., 2004). Wnt5a stimulation

has been shown to enhance the tyrosine kinase activity of ROR2 (Liu et al., 2007;

Akbarzadeh et al., 2008; Liu et al., 2008). Our data confirm previous reports that ROR2

phosphorylation is induced by Wnt5a only and not by Wnt3a (Figure 5E). The loss of correct

ROR2 phosphorylation upon Wnt5a stimulation in Tmem67-/- cells (Figure 5E) therefore

suggests that TMEM67 is essential for the initiation of phosphorylation, possibly by

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

mediating homodimerization. TMEM67 therefore appears to be a receptor of non-canonical

Wnt signalling that preferentially binds Wnt5a with the extracellular cysteine-rich domain

(CRD), and mediates downstream signalling through ROR2 as a co-receptor.

In the present report, lung hypoplasia in Tmem67-/- was dependent on non-canonical

Wnt signalling downstream of Wnt5a/ROR2, for which TMEM67 appeared to be essential

for signalling responses in the developing lung (Figure 6). This is consistent with the

previous finding that non-canonical Wnt5a signalling is essential for proper lung

development through controlling epithelial branching (Li et al., 2002). Defects in lung

branching morphogenesis and the orientation of mitotic divisions in Tmem67-/- ex vivo

cultured lungs were rescued by treatment with the RhoA activator, calpeptin (Figure 7A-C,

Suppl. Figure 5). This confirms previous reports that RhoA activation is essential for

accelerated branching in the developing lungs (Moore et al., 2002; Cloutier et al., 2010;

Moore et al., 2005).

Non-canonical Wnt signalling downstream of Wnt5a was down-regulated in Tmem67-

/- lungs (Figure 6C). However, this was accompanied by increased expression of Shh

transcripts, and downstream effectors of both the Shh pathway (Gli1 and Ptch1) and

canonical Wnt signalling (Axin2; Figure 6D), indicating up-regulation of both the canonical

Wnt and Shh pathways. This is consistent with a previous report that Wnt5a signalling is

essential for inhibition of Shh signalling in the developing lungs after mid-gestation (Li et al.,

2005). This may also explain the greater de-regulation of Wnt signalling compared to Shh

signalling at the mid-gestational time point (E15.5) that we assayed for transcript expression

(Figure 6D). In Tmem67-/-, we therefore suggest that the loss of TMEM67 prevents Wnt5a-

mediated inhibition of Shh signalling in the mutant lungs (Figure 7D). Interestingly, a similar

pulmonary phenotype to Tmem67-/- is observed after ectopic over-expression of Shh in the

developing murine lung after mid-gestation periods (Bellusci et al., 1997). Increased Axin2

expression in Tmem67-/- mutant lungs could similarly be explained by the lack of an

inhibitory effect of the non-canonical Wnt5a ligand on canonical Wnts, since both functional

classes of Wnts have been shown previously to competitively inhibit binding to their receptor

site (Grumolato et al., 2010). This model is also consistent with our previous in vitro results

in Tmem67-/- cells (Abdelhamed et al., 2013). We therefore propose a model in which

signalling through the Wnt5a-TMEM67-ROR2 axis normally represses both Shh and

canonical Wnt signalling (Figure 7D). Loss or mutation of any component in this axis causes

Shh and canonical Wnt signalling de-regulation and ectopic expression, contributing to the

pulmonary hypoplasia, condensed mesenchyme and impaired development of the alveolar

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

system observed in the ciliopathy disease state. Targeting the Wnt5a-TMEM67-ROR2

signalling axis downstream of the receptor site could therefore provide the potential basis for

therapeutic intervention to reduce or prevent lung hypoplasia in ciliopathies.

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

Materials and Methods

Ethics statement

The animal studies described in this paper were carried out under the guidance issued by the

Medical Research Council in Responsibility in the Use of Animals for Medical Research (July

1993) in accordance with UK Home Office regulations under the Project Licence no.

PPL40/3349.

Animals

B6;129P2-Tmem67tm1Dgen/H heterozygous knock-out mice were derived from a line generated

by Deltagen Inc. (San Mateo, CA, USA) and made available from MRC Harwell through the

European Mutant Mouse Archive http://www.emmanet.org/ (strain number EM:02370). The

targeting β-Gal-neo (“geo”) construct inserts downstream of exon one of the Tmem67 gene

(Abdelhamed et al., 2013). Genotyping was done by PCR on DNA extracted from tail tips or

the yolk sac of E11.5-E15.5 embryos, or ear biopsies of adult mice.

Cells

Human embryonic kidney (HEK293) and mouse inner medullary collecting duct (mIMCD3)

cells were grown in Dulbecco’s minimum essential medium (DMEM)/Ham’s F12

supplemented with 10% foetal calf serum at 37 C/5% CO2, essentially as described

previously (Abdelhamed et al., 2013). The derivation and culture of mouse embryonic

fibroblasts (MEFs) has been described previously (Adams et al., 2012) MEFs were grown in

DMEM/Ham’s F12 supplemented with 10% foetal calf serum and 1% penicillin streptomycin

at 37 C/5% CO2.

Cloning, plasmid constructs and transfections. Full-length human TMEM67/MKS3 was

cloned into the pCMV-HA vector as described previously (Adams et al., 2012). The pSec2A-

TMEM67-Nt construct (encoding amino acids F39-T478, and including the cysteine-rich

domain and β-sheet motifs, Figure 4A) was constructed by standard sub-cloning of a PCR

product containing HindIII and NotI restriction sites after amplification with "Platinum" Taq

DNA Polymerase High Fidelity (Life Technologies Ltd., Paisley, UK). Inserts were verified

by bidirectional DNA sequencing. Missense mutations were introduced using the

QuickChange mutagenesis kit (Stratagene Inc., La Jolla, CA, USA) and verified by DNA

sequencing. Plasmid pEF1a-mROR2WT (Mikels et al., 2009) was obtained from Addgene,

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

Cambridge, MA, USA (plasmid number 22613). For transfection with plasmids, cells at 80%

confluency were transfected using Lipofectamine 2000 (Life Technologies Ltd.) according to

the manufacturer’s instructions and as described previously (Dawe et al., 2009).

Antibodies and fluorescent markers

The following primary antibodies were used: mouse anti-β actin (clone AC-15; Abcam Ltd.,

Cambridge, UK); mouse anti-Ki67 (Merck Millipore Inc., Feltham, UK); mouse anti-FLAG

(clone M2; Sigma-Aldrich Co. Ltd., Gillingham, UK); rabbit polyclonal anti-Vangl2 (1:500;

a kind gift from Mireille Montcouquiol, INSERM Université Bordeaux, France); rabbit

polyclonal anti-Gαi3 (1:400; G4040, Sigma Aldrich); rabbit polyclonal anti-atypical Protein

Kinase C (PKC-ζ; 1:400; sc216, Santa Cruz); goat anti-ROR2 (R&D Systems Inc.,

Minneapolis, MN, USA); guinea pig anti-RPGRIP1L (SNC040) polyclonal antibody at 1:200

(Arts et al. 2007), a kind gift from Ronald Roepman, Radboud UMC, Nijmegen, the

Netherlands; and rabbit anti-TMEM67 C-terminus polyclonal antibody at 1:100

(Abdelhamed et al., 2013). Microtubules were stained with mouse monoclonal anti-

acetylated- -tubulin antibody (clone 6-11B-1; Sigma-Aldrich Co. Ltd; 1:1000), shown

previously to detect cochlear ciliary axonemes (Jagger et al., 2011; May-Simera et al., 2009).

Ciliary basal bodies were immunolocalized using a rabbit polyclonal anti-ALMS1 antibody at

1:200 (Jagger et al., 2011). F-actin was stained with tetramethyl-rhodamine (TRITC)-

conjugated phalloidin (Sigma-Aldrich Co. Ltd.) at 1:1000. Secondary antibodies were Alexa-

Fluor-conjugated goat anti-mouse IgG and goat anti-rabbit IgG (Life Technologies Ltd.)

Preparation of tissue sections, histology and immunohistochemistry

Mouse embryos or dissected tissues were fixed in 4% (w/v) para-formaldehyde and

embedded in paraffin wax. Thin sections (4μm) were cut onto “Superfrost Plus” slides (VWR

International Ltd., Lutterworth, UK) and were deparaffinised and rehydrated by standard

methods. Sections were stained with haematoxylin and eosin (VWR International Ltd.) for 2

minutes, then dehydrated in ethanol, cleared in xylene and mounted in DPX. For

immunohistochemistry, tissue sections were deparaffinised and rehydrated. Epitope recovery

was obtained by boiling in 1mM EDTA pH8.0, for 2min using pressure cooker, followed by

20min cooling. Blocking and application of primary antibodies was as described (Dawe et al.,

2007). Appropriate HRP-conjugated secondary antibodies (Dako UK Ltd., Ely, UK) were

used (final dilutions of x10000-25000). Sections were developed in “Sigma Fast” 3,3’-

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

diaminobenzidine (DAB) with CoCl2 enhancer and counterstained with Mayer's

haematoxylin (Sigma-Aldrich Co. Ltd).

Cochlear immunofluorescence and confocal microscopy

For TMEM67 immunofluorescence experiments cochleae were fixed using 2%

paraformaldehyde (PFA) in phosphate buffered saline (PBS) for 20 minutes at room

temperature. For morphogenesis studies cochleae were fixed using 4% PFA in PBS overnight

at 4°C. The organs of Corti were dissected, and divided into 2-3 lengths for subsequent

mounting. Tissues were permeabilized and blocked (0.1% Triton-X 100 with 10% normal

goat serum in PBS) for 30 minutes at room temperature and then incubated in primary

antibodies overnight at 4°C. Following several PBS washes they were incubated in Alexa-

Fluor tagged secondary antibodies (Life Technologies Ltd.) in the dark for 30 minutes at

room temperature. Cells or tissues were mounted on glass slides using Vectashield with

diamidino-2-phenylindole (DAPI; Vector Laboratories Ltd., Peterborough, UK). Imaging was

carried out using a laser scanning confocal microscope (LSM510; Carl Zeiss Microscopy

GmbH, Jena, Germany) or a Nikon Eclipse TE2000-E system, controlled and processed by

EZ-C1 3.50 (Nikon UK Ltd., Kingston-upon-Thames, UK) software. Images were assembled

using Adobe Illustrator CS4 (Adobe Systems Inc., San Jose, CA, USA).

Whole cell extract preparation, western immunoblotting and RhoA activation assays

Whole cell extracts (WCE) containing total soluble proteins were prepared from confluent

untransfected HEK293 or IMCD3 cells, or cells that had been transiently transfected with 1.0

μg plasmid constructs in 90mm tissue culture dishes, or scaled down as appropriate. Ten g

WCE total soluble protein was analysed by SDS-PAGE (using 4-12% polyacrylamide

gradient gels) and western blotting according to standard protocols using either rabbit

polyclonal antisera (final dilutions of 1:200-1000) or mAbs (1:1000-5000). Appropriate

HRP-conjugated secondary antibodies (Dako UK Ltd.) were used (final dilutions of 1:10000-

25000) for detection by the enhanced chemiluminescence “Femto West” western blotting

detection system (Thermo Fisher Scientific Inc., Rockford, IL, USA) and visualized using a

ChemiDoc MP imaging system (BioRad Inc., Hercules, CA, USA). The activated GTP-

bound isoform of RhoA was specifically assayed in pull-down assays using a GST fusion

protein of the Rho effector rhotekin (Cytoskeleton Inc., Denver, CO, USA), using conditions

recommended by the manufacturer. WCEs were processed as rapidly as possible at 4°C, and

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

snap-frozen in liquid nitrogen. Total RhoA (in input WCEs) and pull-down protein was

immunodetected on western blots using a proprietary anti-RhoA monoclonal antibody

(Cytoskeleton Inc.) Immunoblotting for total RhoA was used as the loading control. Ratios of

active RhoA : total RhoA were calculated by quantitating band intensity using ImageLab

5.2.1 software (BioRad Inc.)

Canonical Wnt activity (TOPFlash) luciferase assays

For luciferase assays of canonical Wnt activity, we grew mouse embryonic fibroblasts in 12-

well plates and co-transfected with 0.5 μg TOPFlash firefly luciferase construct (or

FOPFlash, as a negative control); 0.5 μg of expression constructs (pCMV HA-TMEM67, or

empty pCMV-HA/pCMV c myc vector); and 0.05 μg of pRL-TK (Promega Corp., Madison,

WI, USA); Renilla luciferase construct used as an internal control reporter). Cells were

treated with Wnt3a- or Wnt5a-conditioned media to stimulate or inhibit the canonical Wnt

pathway. We obtained Wnt3a- or Wnt5a-conditioned media from stably-transfected L cells

with Wnt3a or Wnt5a expression vectors, and used as described (Willert et al., 2003). Control

media was from untransfected L cells. Activities from firefly and Renilla luciferases were

assayed with the Dual-Luciferase Reporter Assay system (Promega Corp.) on a Mithras

LB940 (Berthold Technologies GmbH & Co. KG, Bad Wildbad, Germany) luminometer.

Minimal responses were noted with co-expression of the FOPFlash negative control reporter

construct. Raw readings were normalized with Renilla luciferase values. Results reported are

from at least four independent biological replicates.

Protein expression and in vitro binding assay

Purified recombinant Wnt3a and Wnt5a proteins (R&D Systems Inc.) and purified BSA as a

negative control (Sigma-Aldrich Co. Ltd.), were labelled with NHS-fluorescein (Thermo

Fisher Scientific Inc.), as described by the manufacturer. Unincorporated fluorescein was

removed by fluorescent dye removal columns (Thermo Fisher Scientific Inc.) TMEM67-Nt

protein (encoding amino acids F39-T478, predicted molecular weight 48kDa) was expressed

following transfection of HEK293 cells with pSec2A constructs (Life Technologies Ltd.)

using conditions recommended by the manufacturer. TMEM67-Nt proteins were diluted in

100 mM bicarbonate/carbonate buffer pH9.6 and applied to “Immunosorb” 96-well plates

(Thermo Fisher Scientific Inc.) overnight at 4°C, washed with 1xPBS, and blocked with 5%

[w/v] non-fat dried milk in 1xPBS for 2 hour at room temperature. Fluorescently-labelled

proteins in blocking buffer were applied to plate wells, incubated for 2 hour at room

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

temperature, and then washed extensively with 1xPBS. Fluorescence retained on plates was

then detected with a Mithras LB940 (Berthold Technologies GmbH & Co. KG) fluorimeter.

Embryonic lung ex vivo culture

Embryonic E12.5 lungs were micro-dissected into cold HBSS (Life Technologies Ltd.) under

completely aseptic conditions. Lungs were washed in serum-free medium and transferred to a

semipermeable transparent “Transwell” membrane with 0.4 µm pore size (Merck Millipore

Inc). The insert was placed over 1 ml of serum-free DMEMF12 medium, supplemented with

penicillin, streptomycin and ascorbic acid (0.2 mg/ml) in a twelve-well plate.

Quantitative Real Time-PCR (qRT-PCR)

qRT-PCR reactions were performed as described previously (Abdelhamed et al., 2013).

Primer sequences are available upon request. The average Ct values of the samples were

normalised to values for β-actin. Fold-difference in expression of the different genes in the

mutant embryos was calculated relative to their expression in wild-type or heterozygous

littermates using the standard curve method.

Measurements and statistical analyses

Length and orientation measurements were carried out using LSM510 Image Browser 4.2

software (Carl Zeiss Microscopy GmbH). Normal distribution of data was confirmed using

the Kolmogorov-Smirnov test (GraphPad Prism, GraphPad Software Inc., La Jolla, CA,

USA). Pairwise comparisons were analysed with Student's two-tailed t-test using InStat

(GraphPad Software Inc.) Results reported are from at least three independent biological

replicates.

Acknowledgements

We thank A. Monk, K. Passam and T. Simpson of Nikon UK Ltd. for technical support and

advice on confocal microscopy. We are very grateful to D. Evans, J. Bilton, C. McCartney

and M. Reay for technical support. We thank R. T. Moon, University of Washington, for the

TOPFlash and FOPFlash constructs. The pEF1a-mROR2WT plasmid was a gift from R.

Nusse, Stanford University School of Medicine, CA, USA. The anti-Vangl2 antibody was a

kind gift from Mireille Montcouquiol, INSERM Université Bordeaux, France. The anti-

RPGRIP1L antibody was a kind gift from Ronald Roepman, Radboud UMC, Nijmegen, the

Netherlands.

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

Competing interests statement

The authors declare that they have no competing conflicts of interests.

Author contributions

Z.A.A., C.A.J. and D.J.J. conceived and designed the experiments. Z.A.A., S.N., C.A.J. and

D.J.J. performed the experiments. All authors analyzed the data and edited the manuscript.

Z.A.A., C.A.J. and D.J.J. wrote the paper.

Funding

We acknowledge funding from the UK Medical Research Council (CAJ; project grant

G0700073), an Egyptian Government Scholarship (ZAA) and a Kid’s Kidney Research

project grant (CAJ and CI). ZAA was supported by a grant from the Rosetree’s Trust (No.

JS16/M279). The research also received funding from the European Community's Seventh

Framework Programme FP7/2009 under grant agreement no: 241955 SYSCILIA. Access to

the B6;129P2-Tmem67tm1Dgen/H line was funded by the Wellcome Trust Knock-out Mouse

Resource scheme (CAJ and CI; grant ME041596). The funders had no role in study design,

data collection and analysis, decision to publish, or preparation of the manuscript.

References

Abdelhamed, Z.A., Wheway, G., Szymanska, K., Natarajan, S., Toomes, C., Inglehearn,

C. and Johnson, C.A. (2013). Variable expressivity of ciliopathy neurological

phenotypes that encompass Meckel–Gruber syndrome and Joubert syndrome is

caused by complex de-regulated ciliogenesis, Shh and Wnt signalling defects. Hum.

Mol. Genet. 22, 1358-1372.

Adams, M., Smith, U.M., Logan, C.V. and Johnson, C.A. (2008). Recent advances in the

molecular pathology, cell biology and genetics of ciliopathies. J. Med. Genet. 45, 257-

267.

Adams, M., Simms, R.J., Abdelhamed, Z., Dawe, H.R., Szymanska, K., Logan, C.V.,

Wheway, G., Pitt, E., Gull, K., Knowles, M.A. et al. (2012). A meckelin–filamin A

interaction mediates ciliogenesis. Hum. Mol. Genet. 21, 1272-1286.

Akbarzadeh, S., Wheldon, L.M., Sweet, S.M.M., Talma, S., Mardakheh, F.K. and

Heath, J.K. (2008). The deleted in brachydactyly B domain of ROR2 is required for

receptor activation by recruitment of Src. PLoS One. 3, e1873.

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

Arts, H.H., Doherty, D., van Beersum, S.E., Parisi, M.A., Letteboer, S.J., Gorden, N.T.,

Peters, T.A., Märker, T., Voesenek, K., Kartono, A. et al. (2007) Mutations in the

gene encoding the basal body protein RPGRIP1L, a nephrocystin-4 interactor, cause

Joubert syndrome. Nat. Genet. 39, 882-888.

Bellusci, S., Furuta, Y., Ruch, M.G., Henderson, R., Winnier, G. and Hogan L.M. (1997)

Involvement of Sonic hedgehog (Shh) in mouse embryonic lung growth and

morphogenesis. Development. 124, 53-63.

Chih, B,, Liu, P., Chinn, Y., Chalouni, C., Komuves, L.G., Hass, P.E., Sandoval, W. and

Peterson, A.S. (2011). A ciliopathy complex at the transition zone protects the cilia as

a privileged membrane domain. Nat. Cell. Biol. 14, 61-72.

Cloutier, M., Tremblay, M. and Piedboeuf, B. (2010). ROCK2 is involved in accelerated

fetal lung development induced by in vivo lung distension. Pediatr. Pulmonol. 45,

966-976.

Consugar, M., Kubly, V., Lager, D., Hommerding, C., Wong, W., Bakker, E., Gattone,

V., II, Torres, V., Breuning, M. and Harris, P. (2007). Molecular diagnostics of

Meckel–Gruber syndrome highlights phenotypic differences between MKS1 and

MKS3. Hum. Genet. 121, 591-599.

Corbit, K.C., Shyer, A.E., Dowdle, W.E., Gaulden, J., Singla, V. and Reiter, J.F. (2008).

Kif3a constrains -catenin-dependent Wnt signalling through dual ciliary and non-

ciliary mechanisms. Nat. Cell Biol. 10, 70-76.

Dabdoub, A. and Kelley, M.W. (2005). Planar cell polarity and a potential role for a Wnt

morphogen gradient in stereociliary bundle orientation in the mammalian inner ear. J.

Neurobiol. 64, 446-457.

Dawe, H.R., Smith, U.M., Cullinane, A.R., Gerrelli, D., Cox, P., Badano, J.L., Blair-

Reid, S., Sriram, N., Katsanis, N., Attie-Bitach, T. et al. (2007). The Meckel–

Gruber Syndrome proteins MKS1 and meckelin interact and are required for primary

cilium formation. Hum. Mol. Genet. 16, 173-186.

Dawe, H.R., Adams, M., Wheway, G., Szymanska, K., Logan, C.V., Noegel, A.A., Gull,

K. and Johnson, C.A. (2009). Nesprin-2 interacts with meckelin and mediates

ciliogenesis via remodelling of the actin cytoskeleton. J. Cell Sci. 122, 2716-2726.

Devenport, D. and Fuchs, E. (2008). Planar polarization in embryonic epidermis

orchestrates global asymmetric morphogenesis of hair follicles. Nat. Cell Biol. 10,

1257-1268.

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

Ezan, J., Lasvaux, L., Gezer, A., Novakovic, A., May-Simera, H., Belotti, E., Lhoumeau,

A.-C., Birnbaumer, L., Beer-Hammer, S., Borg, J.-P. et al. (2013). Primary cilium

migration depends on G-protein signalling control of subapical cytoskeleton. Nat. Cell

Biol. 15, 1107-1115.

Garcia-Gonzalo, F., Corbit, K., Sirerol-Piquer, M., Ramaswami, G., Otto, E., Noriega,

T., Seol, A., Robinson, J., Bennett, C., Josifova, D. et al. (2011). A transition zone

complex regulates mammalian ciliogenesis and ciliary membrane composition. Nat.

Genet. 43, 776-784.

Gerdes, J.M., Liu, Y., Zaghloul, N.A., Leitch, C.C., Lawson, S.S., Kato, M., Beachy,

P.A., Beales, P.L., DeMartino, G.N., Fisher, S. et al. (2007). Disruption of the basal

body compromises proteasomal function and perturbs intracellular Wnt response. Nat.

Genet. 39, 1350-1360.

Green, J.L., Kuntz, S.G. and Sternberg, P.W. (2008). Ror receptor tyrosine kinases:

orphans no more. Trends Cell Biol. 18, 536-544.

Grumolato, L., Liu, G., Mong, P., Mudbhary, R., Biswas, R., Arroyave, R.,

Vijayakumar, S., Economides, E.N. and Aaronson, S.A. (2010). Canonical and

noncanonical Wnts use a common mechanism to activate completely unrelated

coreceptors. Genes Dev. 24, 2517-2530.

Huang, L., Szymanska, K., Jensen Victor, L., Janecke Andreas, R., Innes, A., Davis

Erica, E., Frosk, P., Li, C., Willer Jason, R., Chodirker Bernard, N. et al. (2011).

TMEM237 is mutated in individuals with a Joubert Syndrome related disorder and

expands the role of the TMEM family at the ciliary transition zone. Am. J. Hum.

Genet. 89, 713-730.

Huangfu, D.W., Liu, A.M., Rakeman, A.S., Murcia, N.S., Niswander, L. and Anderson,

K.V. (2003). Hedgehog signalling in the mouse requires intraflagellar transport

proteins. Nature. 426, 83-87.

Jagger, D., Collin, G., Kelly, J., Towers, E., Nevill, G., Longo-Guess, C., Benson, J.,

Halsey, K., Dolan, D., Marshall, J. et al. (2011). Alström Syndrome protein ALMS1

localizes to basal bodies of cochlear hair cells and regulates cilium-dependent planar

cell polarity. Hum. Mol. Genet. 20, 466-481.

Jessen, J.R., Topczewski, J., Bingham, S., Sepich, D.S., Marlow, F., Chandrasekhar, A.,

Solnica-Krezel, L. (2002). Zebrafish trilobite identifies new roles for Strabismus in

gastrulation and neuronal movements. Nat. Cell. Biol. 4, 610-615.

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

Jones, C. and Chen, P. (2007). Planar cell polarity signaling in vertebrates. BioEssays. 29,

120-132.

Jones, C., Roper, V.C., Foucher, I., Qian, D., Banizs, B., Petit, C., Yoder, B.K. and

Chen, P. (2008). Ciliary proteins link basal body polarization to planar cell polarity

regulation. Nat. Genet. 40, 69-77.

Kani, S., Oishi, I., Yamamoto, H., Yoda, A., Suzuki, H., Nomachi, A., Iozumi, K.,

Nishita, M., Kikuchi, A., Takumi, T. et al. (2004). The receptor tyrosine kinase

Ror2 associates with and is activated by casein kinase I epsilon. J. Biol. Chem. 279,

50102-50109.

Khaddour, R., Smith, U., Baala, L., Martinovic, J., Clavering, D., Shaffiq, R., Ozilou,

C., Cullinane, A., Kyttala, M., Shalev, S. et al. (2007). Spectrum of MKS1 and

MKS3 mutations in Meckel syndrome: a genotype-phenotype correlation. Hum.

Mutat. 28, 523-524.

Kiyohashi, K., Kakinuma, S., Kamiya, A., Sakamoto, N., Nitta, S., Yamanaka, H.,

Yoshino, K., Fujiki, J., Murakawa, M., Kusano-Kitazume, A. et al. (2013). Wnt5a

signaling mediates biliary differentiation of fetal hepatic stem/progenitor cells in

mice. Hepatology. 57, 2502-2513.

Lancaster, M.A., Schroth, J. and Gleeson, J.G. (2011). Subcellular spatial regulation of

canonical Wnt signalling at the primary cilium. Nat. Cell Biol. 13, 700-707.

Leightner, A.C., Hommerding, C.J., Peng, Y., Salisbury, J.L., Gainullin, V.G.,

Czarnecki, P.G., Sussman, C.R. and Harris, P.C. (2013). The Meckel syndrome

protein meckelin (TMEM67) is a key regulator of cilia function but is not required for

tissue planar polarity. Hum. Mol. Genet. 22, 2024-2040.

Li, C., Xiao, J., Hormi, K., Borok, Z. and Minoo, P. (2002). Wnt5a participates in distal

lung morphogenesis. Dev. Biol. 248, 68-81.

Li, C., Hu, H., Xioa, J., Chen, H., Li, J.T., Bellusci, S., Delanghe, S. and Minoo, P.

(2005). Wnt5a regulates Shh and Fgf10 during lung development. Dev. Biol. 278, 86-

79.

Liu, Y., Ross, J.F., Bodine, P.V.N. and Billiard, J. (2007). Homodimerization of Ror2

tyrosine kinase receptor induces 14-3-3 beta phosphorylation and promotes osteoblast

differentiation and bone formation. Mol. Endocrinol., 21, 3050-3061

Liu, Y., Rubin, B., Bodine, P.V.N. and Billiard, J. (2008). Wnt5a induces

homodimerization and activation of Ror2 receptor tyrosine kinase. J. Cell.

Biochem.105, 497-502.

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

May-Simera, H.L., Ross, A., Rix, S., Forge, A., Beales, P.L. and Jagger, D.J. (2009).

Patterns of expression of Bardet-Biedl syndrome proteins in the mammalian cochlea

suggest noncentrosomal functions. J. Comp. Neurol. 514, 174-188.

Mikels, A., Minami, Y. and Nusse, R. (2009). Ror2 receptor requires tyrosine kinase

activity to mediate Wnt5A signaling. J. Biol. Chem. 284, 30167-30176.

Montcouquiol, M., Rachel, R.A., Lanford, P.J., Copeland, N.G., Jenkins, N.A. and

Kelley, M.W. (2003). Identification of Vangl2 and Scrb1 as planar polarity genes in

mammals. Nature. 423, 173-177.

Moore, K.A., Huang, S., Kong, Y.P., Sunday, M.E. and Ingber, D.E. (2002). Control of

embryonic lung branching morphogenesis by the Rho activator, cytotoxic necrotizing

factor 1. J. Surg. Res. 104, 95-100.

Moore, K.A., Polte, T., Huang, S., Shi, B., Alsberg, E., Sunday, M.E. and Ingber, D.E.

(2005). Control of basement membrane remodeling and epithelial branching

morphogenesis in embryonic lung by Rho and cytoskeletal tension. Dev. Dyn. 232,

268-281.

Nauli, S.M., Alenghat, F.J., Luo, Y., Williams, E., Vassilev, P., Lil, X.G., Elia, A.E.H.,

Lu, W.N., Brown, E.M., Quinn, S.J. et al. (2003). Polycystins 1 and 2 mediate

mechanosensation in the primary cilium of kidney cells. Nat. Genet. 33, 129-137.

Nonaka, S., Tanaka, Y., Okada, Y., Takeda, S., Harada, A., Kanai, Y., Kido, M. and

Hirokawa, N. (1998). Randomization of left-right asymmetry due to loss of nodal

cilia generating leftward flow of extraembryonic fluid in mice lacking KIF3B motor

protein. Cell. 95, 829-837.

Oishi, I., Suzuki, H., Onishi, N., Takada, R., Kani, S., Ohkawara, B., Koshida, I.,

Suzuki, K., Yamada, G., Schwabe, G.C. et al. (2003). The receptor tyrosine kinase

Ror2 is involved in non-canonical Wnt5a/JNK signalling pathway. Genes Cells. 8,

645-654.

Oishi, I., Kawakami, Y., Raya, A., Callol-Massot, C. and Belmonte, J.C.I. (2006).

Regulation of primary cilia formation and left-right patterning in zebrafish by a

noncanonical Wnt signaling mediator, duboraya. Nat. Genet. 38, 1316-1322.

Park, T.J., Mitchell, B.J., Abitua, P.B., Kintner, C. and Wallingford, J.B. (2008).

Dishevelled controls apical docking and planar polarization of basal bodies in ciliated

epithelial cells. Nat. Genet. 40, 871-879.

Praetorius, H.A. and Spring, K.R. (2001). Bending the MDCK cell primary cilium

increases intracellular calcium. J. Membr. Biol. 184, 71-79.

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

Qian, D., Jones, C., Rzadzinska, A., Mark, S., Zhang, X., Steel, K.P., Dai, X. and Chen,

P. (2007). Wnt5a functions in planar cell polarity regulation in mice. Dev. Biol. 306,

121-133.

Reiter, J.F., Blacque, O.E. and Leroux, M.R. (2012). The base of the cilium: roles for

transition fibres and the transition zone in ciliary formation, maintenance and

compartmentalization. EMBO Rep. 13, 608-618.

Robinson, D.R., Wu, Y.M. and Lin, S.F. (2000). The protein tyrosine kinase family of the

human genome. Oncogene. 19, 5548-5557.

Ross, A.J., May-Simera, H., Eichers, E.R., Kai, M., Hill, J., Jagger, D.J., Leitch, C.C.,

Chapple, J.P., Munro, P.M., Fisher, S. et al. (2005). Disruption of Bardet-Biedl

syndrome ciliary proteins perturbs planar cell polarity in vertebrates. Nat. Genet. 37,

1135-1140.

Roy, J. and Pal, M. (2013). Meckel Gruber syndrome. J. Clin. Diagn. Res. 7, 2102-2103.

Salonen, R. (1984). The Meckel syndrome: clinicopathological findings in 67 patients. Am.

J. Med. Genet. 18, 671-689.

Sang, L.Y., Miller, J.J., Corbit, K.C., Giles, R.H., Brauer, M.J., Otto, E.A., Baye, L.M.,

Wen, X.H., Scales, S.J., Kwong, M. et al. (2011). Mapping the NPHP-JBTS-MKS

protein network reveals ciliopathy disease genes and pathways. Cell. 145, 513-528.

Schneider, L., Clement, C.A., Teilmann, S.C., Pazour, G.J., Hoffmann, E.K., Satir, P.

and Christensen, S.T. (2005). PDGFR alpha alpha signaling is regulated through the

primary cilium in fibroblasts. Curr. Biol. 15, 1861-1866.

Schoenwaelder, S.M. and Burridge, K. (1999). Evidence for a calpeptin-sensitive protein-

tyrosine phosphatase upstream of the small GTPase Rho. A novel role for the calpain

inhibitor calpeptin in the inhibition of protein-tyrosine phosphatases. J. Biol. Chem.

274, 14359-14367.

Schoenwaelder, S.M., Petch, L.A., Williamson, D., Shen, R., Feng, G.S. and Burridge,

K. (2000). The protein tyrosine phosphatase Shp-2 regulates RhoA activity. Curr.

Biol. 10, 1523-1526.

Simons, M., Gloy, J., Ganner, A., Bullerkotte, A., Bashkurov, M., Kronig, C., Schermer,

B., Benzing, T., Cabello, O.A., Jenny, A. et al. (2005). Inversin, the gene product

mutated in nephronophthisis type II, functions as a molecular switch between Wnt

signaling pathways. Nat. Genet. 37, 537-543.

Sipe, C.W. and Lu, X. (2011). Kif3a regulates planar polarization of auditory hair cells

through both ciliary and non-ciliary mechanisms. Development. 138, 3441-3449.

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

Smith, U., Consugar, M., Tee, L., McKee, B., Maina, E., Whelan, S., Morgan, N.,

Goranson, E., Gissen, P., Lilliquist, S. et al. (2006). The transmembrane protein

meckelin (MKS3) is mutated in Meckel-Gruber syndrome and the wpk rat. Nat.

Genet. 38, 191-196.

Sossin, W.S. (2006). Tracing the evolution and function of the Trk superfamily of receptor

tyrosine kinases. Brain Behav. Evol. 68, 145-156.

Szymanska, K. and Johnson, C.A. (2012). The transition zone: an essential functional

compartment of cilia. Cilia. 1, e10.

Szymanska, K., Berry, I., Logan, C., Cousins, S.R., Lindsay, H., Jafri, H., Raashid, Y.,

Malik-Sharif, S., Castle, B., Ahmed, M. et al. (2012). Founder mutations and

genotype-phenotype correlations in Meckel-Gruber syndrome and associated

ciliopathies. Cilia. 1, e18.

Tissir, F., Qu, Y., Montcouquiol, M., Zhou, L., Komatsu, K., Shi, D., Fujimori, T.,

Labeau, J., Tyteca, D., Courtoy, P. et al. (2010). Lack of cadherins Celsr2 and

Celsr3 impairs ependymal ciliogenesis, leading to fatal hydrocephalus. Nat. Neurosci.

13, 700-707.

Valente, E.M., Logan, C.V., Mougou-Zerelli, S., Lee, J.H., Silhavy, J.L., Brancati, F.,

Iannicelli, M., Travaglini, L., Romani, S., Illi, B. et al. (2010). Mutations in

TMEM216 perturb ciliogenesis and cause Joubert, Meckel and related syndromes.

Nat. Genet. 42, 619-625.

Willert, K., Brown, J.D., Danenberg, E., Duncan, A.W., Weissman, I.L., Reya, T., Yates,

J.R. and Nusse, R. (2003). Wnt proteins are lipid-modified and can act as stem cell

growth factors. Nature. 423, 448-452.

Williams, C.L., Li, C., Kida, K., Inglis, P.N., Mohan, S., Semenec, L., Bialas, N.J.,

Stupay, R.M., Chen, N., Blacque, O.E. et al. (2011). MKS and NPHP modules

cooperate to establish basal body/transition zone membrane associations and ciliary

gate function during ciliogenesis. J. Cell. Biol. 192, 1023-1041.

Ybot-Gonzalez, P., Savery, D., Gerrelli, D., Signore, M., Mitchell, C.E., Faux, C.H.,

Greene, N.D.E. and Copp, A.J. (2007). Convergent extension, planar-cell-polarity

signalling and initiation of mouse neural tube closure. Development. 134, 789-799.

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

Figures

Figure 1: Gross anatomical malformations, laterality defects, cardiac defects and

pulmonary hypoplasia in Tmem67-/- mutant mouse embryos and pups. (A) Upper panels:

whole mount E11.5 embryos showing the earliest sign of laterality defects with inverted tail

turning (arrowhead) in a Tmem67-/- mutant embryo. Whole mount lungs of E15.5 embryos

(middle panels) and P0 pups (lower panels). Tmem67-/- E15.5 mutant embryos had identical

left (L) and right (R) lungs, indicating left lung isomerism. Lobes of the right lung in

Tmem67+/+ are numbered as indicated. (B) Upper panels: H&E stained lung tissue section

showing pulmonary hypoplasia, congested vessels and delayed development of the

pulmonary alveoli in an E18.5 Tmem67-/- embryo. Lower panels: immunohistochemical

staining for Ki-67 in E18.5 lung sections. Scale bars = 40μm. (C) IF microscopy of E14.5

lung tissue sections stained for primary cilia (acetylated -tubulin; red), basal bodies (γ-

tubulin; green) and for nuclei with DAPI (blue). Scale bar = 10μm. Bar graphs show primary

cilia length and number in Tmem67+/+ and Tmem67-/- tissues. Statistical significance of the

pairwise comparison is indicated by ** for p<0.01 and *** for p<0.001 Student two-tailed t-

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

test. Error bars indicate s.e.m. (D) Upper panels: whole mount E15.5 embryo images showing

the generalized delayed development, under-developed limbs (white arrowheads) and

omphalocele (red arrowhead) in Tmem67-/- embryos, with detail of limb dysplasia shown

below. Lower panels: whole mount P1 pups, showing reduced body longitudinal axis in the

Tmem67-/- pups. Scale bars = 1cm. (E) Upper panels: H&E stained horizontal section through

the chest cavity of E12.5 Tmem67+/+ and Tmem67-/- animals showing a ventricular septal

defect (VSD) (arrowhead) in the mutant. Scale bar = 100μm. Lower panels: VSD

(arrowhead) in an E15.5 sagittal heart section; scale bar = 200μm. (F) Horizontal sections

through the thoracic cavity of the Tmem67-/- mutant and wild-type control showing aberrant

lung lobulation, dextrocardia, major cardiac malformation and cardiac oedema or pericardial

effusion (asterisk) in the Tmem67-/- embryo. Scale bar = 100μm. (G) H&E (upper panels) and

IHC (lower panels) stained E18.5 liver tissue sections. H&E sections show a persistent

double-layered ductal plate (black arrowheads) around the portal vein branches (pvb) and

abnormally accumulating cells around the pvb in Tmem67-/- embryos (white arrowheads).

IHC stained liver sections for cytokeratin-19 show a double-layered ductal plate and multiple

bile ducts in Tmem67-/- embryos. A normal bile duct in the Tmem67+/+ section is indicated

(arrowhead). Scale bars = 50μm.

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

Figure 2. Orientation defects in stereociliary hair bundles with uncoupling from

kinocilium and basal body position of hair cells in the organ of Corti of neonatal

Tmem67-/- mice. (A) Cochleae dissected from P0 Tmem67+/+ mice (control, left) were

indistinguishable from those of Tmem67-/- littermates (right). Scale bar = 1 mm. (B) Total

length measurements of phalloidin-stained organ of Corti were not significantly different

between control and mutant animals (n=4 cochleae per genotype). (C) Schematic

representation of cellular architecture of the neonatal organ of Corti. There is a single row of

inner hair cells (ihc) located at the neural edge of the sensory epithelium, and three rows of

outer hair cells (ohc1-3) spanning the abneural portion. The hair cell stereociliary bundles

(red) are regularly oriented, with their vertices pointing towards the abneural pole,

corresponding to an alignment of 0° (denoted by vertical dotted line). A line of alignment to

90° is also shown for reference. Ohc are surrounded by a mosaic of non-sensory supporting

cells, including pillar cells (green) and Deiters’ cells (blue). Primary cilia are represented as

black dots. (D) Confocal projections of P0 Tmem67+/+ organ of Corti mid-turn region (50%

of cochlear length) stained for actin using phalloidin to demarcate stereociliary hair bundles

(blue), acetylated α-tubulin antibody (cilia; red) and TMEM67 (green). TMEM67 decorates

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

the proximal regions of cilia in both hair cell types and supporting cells. The magnified inset

shows TMEM67 ciliary localization in a single outer hair cell (arrow) and an adjacent

Deiters’ cell (arrowhead). Scale bar = 10 µm. (E) On the surface of the basal turn (10-20% of

cochlear length) in the organ of Corti of a P0 Tmem67+/+ mouse (left), there was a regular

arrangement of V-shaped stereociliary ohc hair bundles (phalloidin; red), with kinocilia

(acetylated α-tubulin; green) positioned at the abneural pole (around 0°) of hair cells in all

three rows (arrows; shown in magnified insets). Each kinocilium was in close apposition to

the vertex of each hair bundle. Non-sensory supporting cells were also ciliated (arrowheads).

In a Tmem67-/- littermate (right) kinocilia were often mis-localised from the abneural pole of

the hair cell (arrows; shown in magnified insets), and in these cells the orientation of the hair

bundle was uncoupled from the kinocilium position. Adjacent supporting cells were often not

ciliated (arrowheads). Similar effects were seen in the apical turn region (~70-80% cochlear

length). Cytoskeletal staining of inner pillar cells is indicted by asterisks. Scale bar = 10 µm.

(F) Basal body position and hair bundle orientation were tightly coupled in basal and apical

regions of the Tmem67+/+ organ of Corti (left). Uncoupling of hair bundle orientation from

basal body position was apparent in all hair cell rows, in both basal and apical regions in

Tmem67-/- cochleae (detail indicated by arrows is shown in magnified insets). Scale bar = 10

µm. (G) Scatter plots showing hair bundle orientation versus basal body position for

individual ohc in the basal region (corresponding to ~10-20% of cochlear length)o f a

Tmem67+/+ mouse (left; n = 230) and a Tmem67-/- littermate (right; n = 165). Dashed lines

indicate the position of perfect correlation (Pearson’s coefficient of correlation, r = 1). (H)

Genotype-specific differences in basal body position for individual hair cell rows in basal

(10-20%, left) and apical (70-80%, right) cochlear regions. Average deviations from 0° were

significantly different between the genotypes for all rows (pairwise comparisons are indicated

by * for p<0.001, Student unpaired t-test) in both basal and apical regions. Error bars indicate

s.e.m.

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

Figure 3. Normal planar cell polarity and apical planar asymmetry in the organ of Corti

of neonatal Tmem67-/- mice. Confocal projections of P0 Tmem67+/- (left panels) and

Tmem67-/- (right panels) basal turn organ of Corti (corresponding to 10-20% of cochlear

length) stained for actin to demarcate stereociliary hair bundles and cell borders (red). (A) In

both genotypes, Vangl2 (green) localized to supporting cells at the adherens junction with

hair cells. (B) G i3 (green) is enriched in the lateral “bare zone” on the apical surface of

outer hair cells. (C) aPKC (green) is enriched in the medial/neural compartment on the apical

surface of outer hair cells. Mis-aligned hair bundles in Tmem67-/- cochleae (arrows) are

adjacent to normally expressed Vangl2, or display the normal asymmetric expression of G i3

and aPKC. Scale bars = 10 µm.

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

Figure 4. Non-canonical Wnt signalling defects in Tmem67-/- cells and interaction of

Wnt5a with the TMEM67 N-terminus domain. (A) Schematic diagram of conserved

domains and structural motifs within the TMEM67 protein, comprising a signal peptide

(yellow), a cysteine-rich domain (CRD, orange), regions of β-sheet periodicity (grey), seven

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

predicted transmembrane helices (TM, black) and a coiled-coil domain (CC, blue). Locations

are indicated by amino acid residue (aa), with pathogenic missense mutations highlighted in

red. The approximate locations of the two epitopes used to raise N-terminal (Nt) and C-

terminal (Ct) rabbit polyclonal antibodies (Ab) are indicated. The TMEM67 regions used for

exogenous protein expression are indicated by the grey boxes. (B) TOPFlash assays to

quantify canonical Wnt signalling activity in Tmem67+/+ and Tmem67-/- MEFs, following

treatment with either control L-cell or Wnt3a-conditioned media, as indicated, and

cotransfection with empty vector control, wild-type HA-TMEM67, or HA-TMEM67

containing a series of pathogenic missense mutations. Wild-type HA-TMEM67 rescued de-

regulated canonical Wnt signalling in Tmem67-/- cells, but missense constructs did not. (C)

Tmem67-/- cells had a defective response to Wnt5a, expressed as the ratio of Wnt3a response :

combined response to both Wnt3a and Wnt5a. The correct response to Wnt5a was only

rescued with wild-type HA-TMEM67. Values shown are means of at least four independent

replicates and error bars indicate s.e.m. The statistical significance of the pair-wise

comparisons with wild-type HA-TMEM67 values (#) are represented as * for p<0.05, ** for

p<0.01, and *** is p<0.001, Student two-tailed t-test. (D) Left panel: Coomassie-stained

SDS-PAGE analysis of fluorescently-labelled BSA (F-BSA), Wnt3a (F-Wnt3a) and Wnt5a

(F-Wnt5a) proteins. Molecular weights of protein size standards (kDa) are indicated. Middle

panel: the same gel photographed under UV light to show fluorescent labelling of BSA

control, Wnt3a and Wnt5a proteins. Right panel: expression of TMEM67-Nt proteins

(predicted molecular weight 48 kDa), containing the indicated missense mutations. (E)

Preferential in vitro interaction of F-Wnt5a, but not F-Wnt3a or F-BSA negative control, with

increasing amounts of wild-type TMEM67-Nt. (F) Interaction of F-Wnt5a with wild-type

TMEM67-Nt only, but not TMEM-Nt proteins containing the indicated missense mutations.

Values shown are means of three independent replicates and error bars indicate s.e.m. The

statistical significance of the pair-wise comparisons with wild-type TMEM67-Nt values (#)

are represented as * for p<0.05, and ** for p<0.01, Student two-tailed t-test.

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

Figure 5. The receptor tyrosine kinase-like orphan receptor ROR2 co-localizes and

interacts with TMEM67, and is dependent on this interaction for phosphorylation. (A)

Four colour IF imaging showing that endogenous ROR2 (green) co-localizes with TMEM67

(blue) and RPGRIP1L (red) at the ciliary transition zone. Arrowheads indicate regions shown

in magnified insets. DAPI is pseudocoloured in grey. Scale bar = 10μm. (B) Anti-HA co-

immunoprecipitations (IPs) demonstrating interaction between full-length exogenous HA-

tagged TMEM67 (size 115kDa) and FLAG-tagged ROR2 (size 105kDa). Input whole cell

extracts (WCE) for the indicated transfected constructs are on the left. IP of an irrelevant

protein (HA-tagged MCPH1) was a negative control. Results are shown for immunoblotting

(IB) for anti-FLAG (upper panel) and anti-TMEM67 (lower panel). Non-specific band in IPs

is indicated by the asterisk (*); see Suppl. Figure 6 for full unprocessed images. (C) Upper

panel: IPs demonstrating interaction between FLAG-tagged ROR2 and endogenous

TMEM67. Input WCE is shown on the left, and negative control IPs include a no antibody

(Ab) control and goat (Gt) and rabbit (Rb) irrelevant (irr.) polyclonal antibodies (PAb).

Immunoblotting (IB) for anti-FLAG shows pulldown of FLAG-ROR2 by Gt anti-ROR2 and

Rb anti-TMEM67. Lower panel: IPs with irrelevant protein (FLAG-MCPH1, size 93kDa).

(E) Loss of the active phosphorylated ROR2 isoform (labelled P) in mutant Tmem67-/- cells

following Wnt5a treatment, compared to strong induction of the active isoform (upper band,

as indicated) in wild-type Tmem67+/+ cells. Loading control is for β-actin.

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

Figure 6. Loss of Wnt5a-induced branching morphogenesis during Tmem67-/-

embryonic lung ex vivo organogenesis. (A) Embryonic (E12.5) lungs were explanted and

treated for 0, 6 and 24 hr with either control conditioned medium or medium containing

Wnt5a. Magnified insets (black frames) under high power are shown for 24 hr treatments.

Epithelial branching is significantly induced by Wnt5a in Tmem67+/+ lungs, but this response

is absent in Tmem67-/- lungs. The bar graph quantitates the total number of branches in one

lung for each genotype. Values shown are means of three independent replicates and error

bars indicate s.e.m. The statistical significance of the pair-wise comparisons are represented

as * for p<0.05 and n.s. for non-significant, Student two-tailed t-test. Scale bar = 1mm. (B)

H&E staining of ex vivo cultured embryonic lung sections, showing normal acini (ac) and

mesenchymal tissue (ms, in green) for wild-type Tmem67+/+ lung, and the stimulation of

normal epithelial branching by Wnt5a (green asterisk and arrowheads). In contrast, Tmem67-/-

lungs have abnormal mesenchymal cell condensates (red arrowheads), suggesting defective

epithelial-mesenchymal induction. The red asterisks indicate abnormal bronchiolar

formation; cl indicates the direction of the central lung. Scale bar = 100μm. (C) Rho

activation pull-down assays of whole cell extracts from wild-type Tmem67+/+ and mutant

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

Tmem67-/- embryonic (E15.5) lungs. Total RhoA in input material is shown as the loading

control, with the ratio indicating active : total RhoA levels. A positive control for the assay

(+GTPγS; loading with non-hydrolyzable GTPγS) and a negative control (+GDP; loading

with GDP) are also shown. (D) Quantitative real-time PCR assays of transcript expression

levels in wild-type Tmem67+/+ and mutant Tmem67-/- embryonic (E15.5) lungs for Shh,

downstream effectors of the Shh signalling pathway (Gli1 and Ptch1) and a downstream

effector of the canonical Wnt signalling pathway (Axin2). Levels of transcripts were all

significantly increased in Tmem67-/- embryonic lungs, with the indicated pair-wise

comparisons represented as ** for p<0.01, Student two-tailed t-test for n=3 independent

assays. Error bars indicate s.e.m.

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

Figure 7. Rescue of normal embryonic lung branching morphogenesis and polarity in

mutant Tmem67-/- tissue by ex vivo treatment with the RhoA activator calpeptin. (A)

Embryonic lungs (age E11.5) grown in culture for the indicated times after treatment with

either vehicle control (0.1% DMSO) or calpeptin at final concentration 1unit/ml for 3 hours.

Tmem67-/- lungs had abnormal dilated branches (arrowheads) surrounded by areas of

condensed mesenchyme, in contrast to the fine distal branches visible in Tmem67+/+ lungs.

Calpeptin treatment of mutant Tmem67-/- lungs resulted in more developed branch

development and a general morphology that was similar to the wild-type lungs. Magnified

insets are indicated by the black frames and shown on the right. Scale bar = 1mm. (B) The

bar graph quantitates the total number of terminal branches per lung (total n=3) for each

genotype and treatment condition. The statistical significance of the indicated pair-wise

comparisons is represented by * for p<0.05 and ** for p<0.01, Student two-tailed t-test. Error

bars indicate s.e.m. (C) The polarity of mitotic cell division is rescued by calpeptin treatment

from predominantly parallel (para.) in mutant alveoli to predominantly perpendicular (perp.)

divisions, as observed in wild-type epithelia. The statistical significance of the indicated pair-

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

wise comparisons is represented by *** for p<0.001, chi-squared test, with the total number

of cells counted in 10 fields of view indicated above each bar. Representative examples of

mitotic divisions, visualized by -tubulin (green) and indicated by the fine dotted lines, are

shown on the right. Apical surfaces are highlighted by the broad dotted lines, with asterisks

indicating the alveolar lumen. Scale bar = 20μm. (D) Schematic of a model in which

signalling through the Wnt5a-TMEM67-ROR2 axis normally represses Shh and canonical

Wnt (Wnt3a) signalling to moderate levels (small green arrow) between embryonic ages

E9.5-E11.5. Loss or mutation of any component in this axis (red cross) causes loss of

repression (dashed line) with Shh and canonical Wnt pathway de-regulation and ectopic

expression of Shh at later gestation ages (large red arrow). This contributes to pulmonary

hypoplasia with condensed mesenchyme and impaired development of the alveolar system in

the ciliopathy disease state.

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt

Translational impact

Clinical issue

Mutations in proteins that are structural or functional components of the primary cilium cause

a group of comparatively common human inherited conditions known as ciliopathies. Most

clinical features of ciliopathies, such as renal cystic dysplasia, are well-described. However,

pulmonary hypoplasia is a consistent finding in a perinatal lethal group of skeletal

ciliopathies (the short rib polydactyly syndromes) and may be under-reported in another

severe ciliopathy (Meckel-Gruber syndrome), despite being considered as the leading cause

of death in human Meckel-Gruber syndrome patients.

Results

To determine a possible disease mechanism for pulmonary hypoplasia in ciliopathies, this

study characterizes the Tmem67-/- knock-out mouse model for Meckel-Gruber syndrome and

the function of the TMEM67 protein. Pulmonary hypoplasia is a nearly consistent finding in

Tmem67-/- embryos and pups. The study shows that TMEM67 is a receptor of non-canonical

Wnt signalling that preferentially binds Wnt5a and mediates downstream signalling through

ROR2 as a co-receptor. Previous data and the present study confirm that loss or mutation of

any component in the Wnt5a-TMEM67-ROR2 axis contributes to the pulmonary hypoplasia,

condensed mesenchyme and impaired development of the alveolar system observed in the

ciliopathy disease state. Lung branching morphogenesis in Tmem67-/- ex vivo cultured lungs

is rescued by treatment with calpeptin, an activator RhoA (a downstream effector of the non-

canonical Wnt signalling pathway).

Implications and future directions

Our results provide the first evidence that TMEM67 is a receptor, and implicates the Wnt5a-

TMEM67-ROR2 axis during developmental signalling of many tissues. In particular, this

study emphasizes the importance of downstream effectors of non-canonical Wnt signalling

during lung development, and the dysregulation of this pathway in the ciliopathy disease

state. Targeting these effectors could therefore provide the potential basis for therapeutic

intervention to reduce or prevent pulmonary hypoplasia in ciliopathies, and perhaps other

congenital conditions for which pulmonary hypoplasia is a complication.

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Acce

pted

man

uscr

ipt